Model benchmarks
byteshape/Qwen3.8-27B-GGUF:IQ3_XS local LLM performance
As of November 2026, byteshape/Qwen3.8-27B-GGUF:IQ3_XS runs at up to 45.8 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
27B
Peak speed
45.8 tok/s
Average speed
45.5 tok/s
Avg PP
432.1 tok/s
Min memory
15.8 GB
Max context
128,000 tokens
Avg output / run
46,965 tokens
Avg runtime / run
17m 24s
Avg quality
83.7
Benchmark runs
2
GPUs tested
1
Quality by task
Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 82.7 | 83.7 | 84.7 |
| Agent Workflow | 69.9 | 75.5 | 81.0 |
| Code Generation | 79.9 | 81.2 | 82.5 |
| Role Play & Narrative | 90.4 | 92.3 | 94.1 |
| Research & Analysis | 83.6 | 85.9 | 88.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination byteshape/Qwen3.8-27B-GGUF:IQ3_XS has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon RX 7800 XT | llama.cpp | — | 45.8 tok/s | 45.5 tok/s | 15.8 GB | 128,000 tokens | 83.7 | 2 |
Benchmark runs
All 2 byteshape/Qwen3.8-27B-GGUF:IQ3_XS runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is byteshape/Qwen3.8-27B-GGUF:IQ3_XS good for coding?
- In our benchmarks, byteshape/Qwen3.8-27B-GGUF:IQ3_XS scores 81.2/100 for coding. It runs at about 45.5 tok/s, so if you want more speed, nvidia/Qwen3.8-27B-NVFP4 is faster (~63.3 tok/s) and still scores well for coding (82.7/100).
- Is byteshape/Qwen3.8-27B-GGUF:IQ3_XS good for agentic (tool-using) tasks?
- In our benchmarks, byteshape/Qwen3.8-27B-GGUF:IQ3_XS scores 75.5/100 for agentic workflows. It runs at about 45.5 tok/s, so if you want more speed, swift-iq2_xs is faster (~91.5 tok/s) and still scores well for agentic workflows (91.4/100).
- How fast is byteshape/Qwen3.8-27B-GGUF:IQ3_XS for local inference?
- Across 2 community benchmark runs, byteshape/Qwen3.8-27B-GGUF:IQ3_XS reaches up to 45.8 tok/s and averages 45.5 tok/s, with the fastest results on AMD Radeon RX 7800 XT.
- How much memory does byteshape/Qwen3.8-27B-GGUF:IQ3_XS need?
- The leanest observed configuration used about 15.8 GB of memory.
- Which tools have been used to run byteshape/Qwen3.8-27B-GGUF:IQ3_XS?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.